Here's all that's new on Hyperstack this week:
You can now run inference for the latest OpenAI gpt-oss-120b model directly on AI Studio, making it easier than ever to test and deploy them without additional setup.
Fine-tuning for this model will also be available soon, giving you even more flexibility to customise performance for your specific needs. Check out the pricing below - the gpt-oss model is one of the most affordable option for running inference on AI Studio:
Run your AI workloads on the powerhouse NVIDIA H100 PCIe GPU at a fraction of the cost with Spot VMs. Perfect for training, inference and batch jobs that can handle interruptions, without compromising performance. Get H100 PCIe Spot VMs Now at Just $1.52/hr!
Launch NVIDIA H100 PCIe Spot VM
Logging into Hyperstack now automatically signs you into AI Studio, no extra steps or more logins. Log out once and you’re fully signed out. This ensures secure and seamless authentication across both platforms. If you haven’t explored AI Studio yet, now’s the perfect time to get started. Check out our documentation here to kickstart your journey.
This week, we have also made key updates to improve overall stability and platform flow.
Here's what's coming soon our our cloud GPU platform:
Image support for gpt-oss 120B and 20B is on the way! While you wait, check out our tutorial on running gpt-oss 20B on NVIDIA H100 PCIe GPUs.
The NVIDIA RTX 6000 Pro Blackwell is launching soon on Hyperstack. Reserve early to be the first to get access.
Reserve Your NVIDIA RTX 6000 Pro
Check out our latest blog on Hyperstack:
In our latest blog, we discuss gpt-oss and show how Hyperstack makes running these open models effortless at enterprise scale. For the first time since GPT‑2, OpenAI has dropped fully open‑weight language models that you can download, run and own. The long‑awaited gpt‑oss‑20B and gpt‑oss‑120B are released under the Apache 2.0 license. You get the state‑of‑the‑art AI directly in your hands.
Check out the full blog below!
In our latest tutorial, we deploy gpt-oss‑20B, OpenAI’s powerful open‑weight language model designed for efficient local deployment and advanced reasoning. The model contains approximately 20 billion parameters and uses a mixture‑of‑experts (MoE) architecture. This design allows it to activate only a portion of its parameters for each token, delivering strong performance while keeping compute requirements manageably.
Check out the full blog below!
In our latest blog, we explored the NVIDIA A100 GPU family, covering PCIe, PCIe with NVLink, and SXM configurations. We discuss specs, features, pricing models and reservation process on Hyperstack, helping you choose the right GPU for AI training, inference and HPC. Learn how to optimise performance, manage costs and secure reliable GPU capacity for your projects.
Check out the full blog below!
In our latest blog, we explore the NVIDIA RTX A6000, covering its specs, features, pricing models, and how to reserve your GPU VM on Hyperstack. Learn how snapshots, bootable volumes and flexible storage empower AI, rendering and HPC workloads. Whether for on-demand, reserved or spot VMs, this guide helps you deploy efficiently.
Check out the full blog below!
At Hyperstack, we’re here to support builders, creators and innovators who are pushing AI forward, from early prototypes to production-ready deployments.
Got something exciting in the works? We’d love to hear about it. Stand a chance to be featured in our next Weekly Rundown.
Share your story here.
That's it for this week's Hyperstack Rundown! Stay tuned for more updates next week and subscribe to our newsletter below for exclusive AI and GPU insights delivered to your inbox!
Catch up on everything you need to know from Hyperstack Weekly below: